RPC allows an application programmer to call procedures on different sets of machines without worrying about the network details. Sun RPC (also known as ONC RPC) is the most widely used mechanism by distributed applications, and TI-RPC (also known as ONC+ RPC) is its latest version. This article describes various RPC concepts and explains how to get started creating client and server applications using Sun TI-RPC on the AS/400.
As enterprises become more interconnected, there is an increased need to distribute data across the network. This fact helps drive a continued investment in the development of true production distributed applications. Different applications lend themselves to different models of distributed application development. Generally, three models for distributed application development exist:
• Shared-data, in which the application treats the data as local data and some lower- level procedure handles moving it across the network; in OS/400, DB2/400 multisystem support allows multiple systems to share a single consistent model of a relational database
• Message passing, such as the IBM MQSeries, which functions as a middleware base for distributed applications on the system
• Remote Procedure Call (RPC), which separates and distributes a client application from a server mechanism
With OS/400 V4R2, IBM added support for TI-RPC, the most widely used remote procedure call in the industry. AS/400 TI-RPC provides a solid operating system base for delivering enterprise applications based on the RPC model. In this article, I’ll explain how TI-RPC is implemented in OS/400; what its mechanisms, levels, and concepts are; and how to develop TI-RPC services and applications with your AS/400.
Mechanisms of TI-RPC
Sun Microsystems developed ONC RPC, and distributed applications frequently use ONC RPC to communicate with each other. Transport In-dependent RPC, or TI-RPC
(also known as ONC+ RPC), is the latest versionof ONC RPC. To take advantage of TIRPC, you will need OS/400 V4R2 and above.
RPC is based on extending the notion of conventional, or local, procedure calling so that the called procedure need not exist in the same address space as the calling procedure. The two processes may be on the same system, or they may be ondifferent systems with a network connecting them. Figure 1 shows how PC works.
TI-RPC Concepts
RPC provides a mechanism by which services can register themselves with a daemon (like a server job), called the RPCBIND daemon, which allows clients to access these services. The clients obtain the addresses of these services by querying the RPCBIND daemon by using one of the TI-RPC APIs. As such, the following components provide a complete set of application development tools.
• Authentication Mechanism. The implementation of TI-RPC on the AS/400 provides for two methods of authenticating a client: the default method, which is NULL, or none; and the system authentication method. The system authentication method allows a client to pass such information as hostname, UID, and group membership to the remote service. It is the service’s responsibility to handle this information and reply to the client if the authentication is insufficient or incorrect.
• Network Selection. The Intermediate and Expert Levels of TI-RPC use the Network Selection APIs. The network selection mechanism allows you to choose the transport protocol on which an application should run (OS/400 TI-RPC supports only TCP and User Datagram Protocol, UDP).
• Name-to-Address Translation. The Expert Level of TI-RPC uses the Name-to- Address Translation APIs. They provide an interface similar to the socket calls gethostbyname(), gethostbyaddr(), and getservbyname().
• eXternal Data Representation (XDR). XDR allows you to send a C data type, converted to the correct byte order (if necessary), across the network. On the AS/400, National Language Support (NLS) support has been added to XDR for certain functions, allowing data to be communicated between clients and services running in different code pages.
• RPCBIND Daemon. When it comes to RPC programming, RPCBIND is the most important service. It acts as the directory for RPC applications running on the machine. When a client wants to connect to an RPC service, it contacts the RPCBIND daemon and requests the address. In this way, addresses can be dynamic, and the client is not required to know which port the service is waiting on. Also, for any service to be useful, it must be registered with the RPCBIND daemon. Services will not start up if the RPCBIND daemon is down; neither will clients be able to find the services if the RPCBIND daemon is down.
On the AS/400, use the Start RPC Binder Daemon (RPCBIND) command to start the RPCBIND daemon. Use the End RPC Binder Daemon (ENDRPCBIND) command to end the daemon.
• RPCGEN Compiler. RPC has it’s own pseudo language, called RPCL, to allow for the quick prototyping of RPC applications. By using the RPCGEN compiler, a developer can generate client and server stub files, as well as more complex XDR conversion routines. These stub files can then be modified to finish the development of an application. On the AS/400, the RPCGEN compiler takes stream files stored in an Integrated File System (IFS) directory as input and places output in the stream files as well.
On the AS/400, use the Convert RPC Source (RPCGEN) command to invoke the RPCGEN compiler. OS/400 also specifies that RPCL input source files must reside in the “root” (/) or QOpenSys file system.
Levels of TI-RPC APIs
TI-RPC provides various levels of support based on the need and expertise of the users. The support extends from providing a simple RPC mechanism to providing a set of sophisticated interfaces, as illustrated by the following TI-RPC levels or layers.
• Simplified Level. The simplified level combines all the API calls into one procedure. Although it does not allow customization of the client or the service, you can use the simplified level to develop (or prototype) an RPC service and corresponding client application quickly.
• Top Level. The top level allows more customization to both the client and the service while still maintaining an ease of development and use. The parameters are similar to the simplified level.
• Intermediate Level. The intermediate level is the first layer where the differentiation between layers becomes noticeable. The APIs are identical to the top level, except the developer is responsible for using the Network Selection APIs, allowing more customization and greater control over the transport that is used.
• Expert Level. The expert level is the lowest layer of TI-RPC APIs available on the AS/400. It performs like the intermediate layer, except the developer is responsible for Network Selection and Name-to-Address Translation. The customization of the client and the service is much greater at this layer, allowing control over the transport, the buffer sizes, and other minute details of the application.
You can use other APIs in conjunction with all the layers except the simplified layer. They provide methods for sending back errors from the service to the client, freeing space allocated to the clients and services, and providing enhanced error detection and reporting.
Developing TI-RPC Services
To help you with TI-RPC, I have posted sample code on the MC Web site at www. midrangecomputing.com/mc/99/04 that illustrates the concepts of top-level service APIs you use in developing TI-RPC services, and of top-level client APIs you use in developing TI-RPC client applications. These samples are meant to illustrate TI-RPC coding concepts. You should use these samples as an educational tool, not as a complete product, to learn more about programming in TI-RPC.
The development of a service at the top level requires the developer to write a dispatch routine (also shown in the Web code). At this level, a dispatch routine is called when a service request comes in. The dispatch routine must collect the arguments and call the correct local procedure, catch all errors and results, and return that information to the client. After a dispatch function is written, it can be readily copied and used in other services with only slight modifications.
The top, intermediate, and expert layers can use the same dispatch function without modification. The advantage of the top level over the other lower layers is the ability to specify the network transport type (or nettype) as a string instead of by using the network selection APIs. After calling the top level API, the service is created, bound to the dispatch function, and registered with the RPCBIND service.
For the TI-RPC client application, at the top level, you must create a client handle before you can work with the distributed application. Top level APIs are easy to use, and they allow more manipulation and error handling than the simplified level.
For more detailed information on client and service application development, please see the IBM and AS/400 Partners in Development TI-RPC developers’ Web site at www.as400.ibm.com/developer/tirpc.
ONC RPC vs. ONC+ RPC (TI-RPC)
TI-RPC provides similar functionality as provided by the ONC RPC, but with more flexibility. For example, while creating client or server handles, you do not need to hard code the transport protocol in the application; in other words, a call to clnttcp_create() can be replaced with a call to clnt_tp_create() or clnt_tli_create(), and you can pass the transport protocol as a parameter. Figure 2 lists some of the changed interfaces.
Traditionally, a vast majority of distributed applications have used Sun RPC for communications between the client and the server. With OS/400 Sun TI-RPC support, you can easily port RPC-based applications to the AS/400, allowing AS/400 applications to
communicate transparently with other applications running on UNIX, Windows, OS/390, Linux, and OS/2. In addition, a number of major applications such as Intelligent Miner for Data, Network File System (NFS), and CallPath/400 that have been ported to the AS/400 rely on Sun RPC for communications.
I encourage you to start working with TI-RPC and begin integrating it into your environments. The AS/400 TI-RPC Web site is an excellent place to get started. Between this Web site and those documents listed under “Additional Reading,” you’ll be porting TIRPC applications to your AS/400 in no time.
Reference
IBM AS/400 Partners in Development TI-RPC (ONC+RPC) Web site: www. as400.ibm.com/developer/tirpc
Additional Reading
“Commonsense AS/400 NFS Security Measures,” AS/400 Network Expert, March/ IBM TI-RPC API Guide: www.as400.ibm. com/developer/tirpc/rpcrefman.html IBM RPC Programming Guide: www.as400. ibm.com/developer/tirpc/rpcguide.html
“Share Files with UNIX Systems Through AS/400 NFS,” MC, February 1999 Sun Microsystems ONC+ Developer’s Guide: docs.sun.com/ab2/coll.45.4/ONCDG/ @Ab2TocView/1363
HOST A HOST B
 |
Client Program
Request
Completed
ONC RPC Interfaces ONC+ RPC (TI-RPC) Interfaces
callrpc() rpc_call() registerrpc() rpc_reg() clntudp_create() or clnttcp_create() clnt_tp_create() or clnt_tli_create() svcudp_create() or svctcp_create() svc_tp_create() or svc_tli_create() svc_register() svc_reg()
svc_unregister() svc_unreg()
pmap_set() rpcb_set()
pmap_unset() rpcb_unset()
Figure 2: Here’s how ONC RPC interfaces stack up against TI-RPC.
More than ever, there is a demand for IT to deliver innovation. Your IBM i has been an essential part of your business operations for years. However, your organization may struggle to maintain the current system and implement new projects. The thousands of customers we've worked with and surveyed state that expectations regarding the digital footprint and vision of the company are not aligned with the current IT environment.
TRY the one package that solves all your document design and printing challenges on all your platforms. Produce bar code labels, electronic forms, ad hoc reports, and RFID tags – without programming! MarkMagic is the only document design and print solution that combines report writing, WYSIWYG label and forms design, and conditional printing in one integrated product. Make sure your data survives when catastrophe hits. Request your trial now! Request Now.
Forms of ransomware has been around for over 30 years, and with more and more organizations suffering attacks each year, it continues to endure. What has made ransomware such a durable threat and what is the best way to combat it? In order to prevent ransomware, organizations must first understand how it works.
Disaster protection is vital to every business. Yet, it often consists of patched together procedures that are prone to error. From automatic backups to data encryption to media management, Robot automates the routine (yet often complex) tasks of iSeries backup and recovery, saving you time and money and making the process safer and more reliable. Automate your backups with the Robot Backup and Recovery Solution. Key features include:
Business users want new applications now. Market and regulatory pressures require faster application updates and delivery into production. Your IBM i developers may be approaching retirement, and you see no sure way to fill their positions with experienced developers. In addition, you may be caught between maintaining your existing applications and the uncertainty of moving to something new.
IT managers hoping to find new IBM i talent are discovering that the pool of experienced RPG programmers and operators or administrators with intimate knowledge of the operating system and the applications that run on it is small. This begs the question: How will you manage the platform that supports such a big part of your business? This guide offers strategies and software suggestions to help you plan IT staffing and resources and smooth the transition after your AS/400 talent retires. Read on to learn:
LATEST COMMENTS
MC Press Online